Abstract Superpixel segmentation of 2D image has been widely used in many computer vision tasks. However, limited to the Gaussian imaging principle, there is not a thorough segmentation solution to the ambiguity in defocus and occlusion boundary areas. In this paper, we consider the essential element of image pixel, i.e., rays in the light space, and propose light fifield superpixel (LFSP) segmentation to eliminate the ambiguity. The LFSP is fifirst defifined mathematically and then a refocus-invariant metric named LFSP self-similarity is proposed to evaluate the segmentation performance. By building a clique system containing 80 neighbors in light fifield, a robust refocus-invariant LFSP segmentation algorithm is developed. Experimental results on both synthetic and real light fifield datasets demonstrate the advantages over the state-of-the-arts in terms of traditional evaluation metrics. Additionally the LFSP self-similarity evaluation under different light fifield refocus levels shows the refocus-invariance of the proposed algorithm